Abstract
BACKGROUND: In renal patients estimation of GFR is routinely done by means of population-based formulae using serum creatinine levels. For GFR determination in the creatinine-blind regions or in cases of reno-hepatic syndrome as well as in critical cases of live kidney donors individualized measurements of GFR (mGFR) employing the kinetics of exogenous filtration markers such as the inulin-like polyfructosan sinistrin are necessary. The goal of this study is to compare mGFR values with the eGFR values gained by the Modification of Diet in Renal Disease (MDRD4) and Chronic Kidney Disease-Epidemiology Collaboration (CKD-EPI) formulae. METHODS: In 170 subjects comprising persons with normal renal function or with various stages of kidney diseases (CKD 1-4) GFR was measured by application of intravenous bolus of sinistrin and assessment of temporal plasma concentration profiles by means of pharmacokinetic methods (mGFR). Comparisons of mGFR with MDRD4- and CKD-EPI-derived eGFR values were performed by means of linear regression and Bland-Altman analyses. RESULTS: Reasonable agreement of mGFR and eGFR values was observed in patients with poor renal function [GFR below 60 (ml/min)/1.73 m²]. In cases of normal or mildly impaired renal function, GFR determination by MDRD4 or CKD-EPI tends to underestimate GFR. Notably, there is practically no difference between the two eGFR methods. CONCLUSIONS: For routine purposes or for epidemiological studies in cases of poor renal function eGFR methods are generally reliable. But in creatinine-blind ranges [GFR above 60 (ml/min)/1.73 m²] eGFR values are unreliable and should be replaced by clinically and physiologically suitable methods for mGFR determination. CONSORT: http://www.consort-statement.org/index.aspx?o=1190.